The Joint Commission Journal on Quality and Patient Safety 2019; 000:1–10

Use of Systems Engineering to Design a Hospital Command Center

Erin M. Kane, MD; James J. Scheulen, PA, MBA; Adrian Püttgen, MD; Diego Martinez, PhD; Scott Levin, PhD; Bree A. Bush; Linda Huffman, DNP, RN; Mary Margaret Jacobs, MS, RN; Hetal Rupani, MHA; David T. Efron, MD

Background: In hospitals and health systems across the country, patient flow bottlenecks delay care delivery—emergency department boarding and operating room exit holds are familiar examples. In other industries, such as oil, gas, and air traffic control, command centers proactively manage flow through complex systems.

Methods: A systems engineering approach was used to analyze and maximize existing capacity in one health system, which

led to the creation of the Judy Reitz Capacity Command Center. This article describes the key elements of this novel health

system command center, which include strategic colocation of teams, automated visual displays of real-time data providing a global view, predictive analytics, standard work and rules-based protocols, and a clear chain of command and guiding tenets. Preliminary data are also shared.

Results: With proactive capacity management, subcycle times decreased and allowed the health system’s flagship hospital to increase occupancy from 85% to 92% while decreasing patient delays.

Conclusion: The command center was built with three primary goals—reducing emergency department boarding, elim- inating operating room holds, and facilitating transfers in from outside facilities—but the command center infrastructure has the potential to improve hospital operations in many other areas.

I

PROBLEM DEFINITION AND CONTEXT

n the inpatient setting, hospital gridlock often delays pa-tient movement to the optimal location for care. Emer-

gency department (ED) crowding and boarding, whichstem from systemwide inefficiencies, 1–3 have been directlylinked to higher inpatient morbidity, preventable harm, andoverall mortality, as well as increased total inpatient lengthof stay and decreased patient satisfaction. 4–16 Operatingroom (OR) exit holds, in which patients are unable to movefrom the OR to the postanesthesia care unit (PACU) orthe ICU due to capacity constraints, are associated withworse patient outcomes and avoidable expense. 17,18 Delaysin moving patients to critical care units are associated withincreased mortality. 19–21 At tertiary care centers, inefficien-cies often limit the number of patients eligible for transferin from surrounding community hospitals, thereby delay-ing or preventing access to care.

Many approaches have been tried to address boardingand crowding, such as monitoring of bed turnaround time,OR schedule smoothing, telemedicine consults, Lean andPlan–Do–Study–Act (PDSA) rapid cycle improvement,and many more. 2,5,22,23 Despite this, crowding continuesto increase, leading The Joint Commission to require hos-pitals to measure and address ED boarding and the Centersfor Medicare & Medicaid Services to include related metrics

1553-7250/$-see front matter © 2018 The Joint Commission. Published by Elsevier Inc. All rights reserved. https://doi.org/10.1016/j.jcjq.2018.11.006

in public reporting. 22,24 New approaches to flow manage-ment are vital to effect and sustain change. 25,26

At the Johns Hopkins Hospital (JHH), we used a sys-tems engineering approach to analyze and maximize exist-ing capacity, which led to the creation of the Judy ReitzCapacity Command Center. To our knowledge, this is thefirst hospital command center of this scale and breadth tobe described in the literature. Previously, hospitals have hadmore limited transfer centers or access centers 27,28 used tofacilitate the movement of patients from outside hospitals,as well as bed management centers with a more comprehen-sive view of total hospital bed allocation. 29,30 More recently,some institutions have described an expanded operationscenter approach to managing patient flow. 31–33 Our workbuilds on these and incorporates many of the approaches toboarding that have been described previously.

Our effort is unique in several key ways. The Capac-ity Command Center (CCC) centralizes previously isolatedadministrative processes and local performance initiatives.This global view is essential to prioritize projects, share bestpractices, and standardize work across the hospital. Further,the CCC is novel in the degree to which it incorporatesreal-time data, predictive analytics, and simulation model-ing through its grounding in systems engineering. Finally,the CCC’s permanence, as signaled by its physical infras-tructure and organizational integration, provides a longitu-dinal and persistent platform for culture change.

To achieve the broader mission and vision of proactivelymanaging patient flow throughout an institution, exampleswere taken from industries outside of health care. Industries

2 Erin M. Kane, MD, et al. Use of Systems Engineering to Design a Hospital Command Center

Table 1. Command Center Teams

Team Function Workload

Hopkins Access Line (HAL) Connects physicians by phone, including for transfer requests, and activates emergency teams

20,000 calls monthly, including 700 transfer requests

Admission Services Processes registration and insurance notification for all patients

170 admissions per day, including

hospitalizations and OR admissions HopComm/Lifeline Transports patients between facilities by air

or ground and performs critical care transport in-hospital

3,000 transports monthly

Bed Management nursing coordinators Assign patients with a pending bed request from the ED, the admitting office, the PACU, or an outside facility to an appropriate inpatient unit

100–130 patients matched to beds daily

Capacity Optimization leadership group Drives strategic and operational initiatives Led by Chief Administrative Officer for Capacity Management Director of Patient/Family and Visitor Services, and Medical Director of Capacity Management

30–40 projects ongoing

OR, operating room; ED, emergency department; PACU, postanesthesia care unit.

such as oil and gas, air traffic control, NASA, the Olympics,and entire city governments have developed command cen-ters to proactively manage operations. 34–38 Core elementsof these command centers that we replicated include strate-gic colocation of teams, automated visual displays of real-time data providing a global view, predictive analytics, stan-dard work and rules-based protocols, and a clear governancestructure and guiding tenets.

Our capacity optimization journey began with three pri-mary goals: (1) expediting admissions from the ED and re-ducing ED boarding, (2) streamlining perioperative flowand eliminating OR holds, and (3) facilitating critical trans-fers from outside facilities. The key elements of the CCC,described below, were developed with these goals and mod-els from other industries in mind.

INITIAL APPROACH

The JHH is the flagship hospital for the six-hospital JohnsHopkins Health System and is a 1,131-bed tertiary carehospital located in Baltimore. In 2017 there were more than65,000 ED visits and 48,000 admissions; 18% of admis-sions were transfers from outside facilities via the HopkinsAccess Line (HAL). Average operational occupancy acrossall inpatient units is 90%, with a 95% average occupancyfor the Department of Medicine.

The Judy Reitz Capacity Command Center is physicallycomposed of 5,500 square feet of space with 38 worksta-tions, three conference rooms, and eight offices surround-ing 22 digital screens with 13 real-time analytic tiles feedingin data from seven different sources ( Figure 1 ). The devel-opment of the CCC required physical, human, and capitalresources. Physical space was allocated in a central locationin the hospital. The majority of the staff came from fourpreexisting departments that were consolidated, and a few

new positions were funded specifically for the CCC. Com-puter hardware and engineering was an upfront cost, andongoing financial support has been built into the hospitalbudget.

Colocation of Key Teams

When a patient is admitted to the JHH, up to four teamsare involved: (1) the HAL, or transfer line that connectsphysicians, (2) Admission Services, (3) Lifeline transport,and (4) Bed Management ( Table 1 ). The process beginswith a request for an admission bed, which may be for ascheduled/planned admission or for an admission via theED or from an outside hospital via HAL. Next, a nursingbed management coordinator identifies an appropriate bedand assigns it to that patient. If no bed is available at thetime of request, the patient is placed in a queue until a bedcan be assigned. After a bed is assigned, Lifeline facilitatestransport in for patients transferred from outside hospitals,and Admission Services processes registration. All patientsare touched by a minimum of two of the four teams.

In designing the CCC, we intentionally colocated theseteams. Literature from other industries, in particular soft-ware engineering, has demonstrated that teams separatedby even 30 meters take 2.5 times as long to complete aproject as a colocated team. 39 Colocation has been studiedand found to be key in a range of industries, including gov-ernment programming and health services. 40–43 The rea-sons for this are thought to be both formal and informalmechanisms used to coordinate work. In one striking ex-ample from the transportation industry, authors highlighthow colocation yields efficiencies: “operators of the LondonUnderground developed ways of working together with-out specifically coordinating their efforts, relying on speak-ing out loud and monitoring.”39 (p. 320) When the personresponsible for scheduling the trains told a train driver to

Volume 000, No. , January 2019 3

Figure 1: The current configuration of the Capacity Command Center includes nursing bed management coordinators, transfer line operators, admitting coordinators, transport coordinators, and drop-in workspace.

slow down, the person responsible for passenger announce-ments responded by announcing overhead that the nexttrain would be delayed. 39 In our case, teams that wouldpreviously communicate by fax, page, or phone now com-municate in person. For instance, the Lifeline operators co-ordinate with the HAL expeditors to dispatch critical caretransport teams.

Real-Time Data

At the time of initiation of the project, data related to pa-tient flow and hospital capacity were distributed in sevenunique systems that, for the most part, did not interfacewith each other. The data could be accessed but with sig-nificant effort and time for each instance. At the heart of theCCC are the “tiles” that project integrated feeds from all rel-evant data sources and provide data in real time. These col-

lective tiles form the Wall of Analytics ®, which was devel-oped in partnership with GE Healthcare ( Figure 2 ). Teamsof engineers, data analysts, administrators, physicians, andnurses designed the tiles sequentially. The tiles are not in-tended to be static or retrospective but rather to promoteimmediate and timely action by displaying real-time andpredictive analytics.

Analytics can be described along a continuum: descrip-tive, diagnostic, predictive, and prescriptive. 44–47 To start,active collection and collation of real-time descriptive dataenhanced our ability to make decisions in a timely manner.As the CCC matures, the tiles have been refined to incor-porate more predictive and prescriptive capabilities. Predic-tive data forecast occupancy and wait times, while prescrip-tive tiles provide decision support for CCC staff, such as arecommended bed placement. The CCC tiles extract data

4 Erin M. Kane, MD, et al. Use of Systems Engineering to Design a Hospital Command Center

Figure 2: There are 14 total tiles on the Wall of Analytics. Tiles update in real time, and initial design required integration

of data streams from seven different systems. Four examples of tiles are included here: the ED status tile, which displays adult and pediatric ED census and boarders along with boarding time; the inbound HAL patient tile, which gives a visual representation of patients awaiting bed assignment and transport (including visual cues for ICU and ED patients); the bed

summary tile, which gives an accounting for bed availability by functional unit and drills down to the unit level when clicked; and the forecast occupancy tile, which uses predictive analytics to project census for the next 48 hours. ED, emergency department; HAL, Hopkins Access Line.

Volume 000, No. , January 2019 5

Figure 2: Continued

from disparate systems, process them, and display them ina way that is digestible and actionable for the end user.

In addition to real-time data feeds, dashboards summa-rizing historical trends on key metrics were developed aspart of the capacity optimization effort. Though not dis-played in the CCC, these have been integral to leadershipmeetings and are used to guide strategic and operationalefforts.

Standard Operating Procedures

Standard operating procedures (SOPs) that emphasize highreliability and low variability were developed and detailedin a written manual. These describe who in the CCC re-sponds to information on the tiles and what action shouldbe taken. They are organized around the three primary goals

of reducing ED boarding, streamlining perioperative flowand eliminating OR holds, and facilitating transfers fromoutside hospitals. For example, one of a series of SOPs thatgoverns perioperative flow describes actions when a patientin the PACU is not cleared within 90 minutes: A red tri-angle that appears on the procedural tile triggers the PACUcharge nurse to conference with the bedside nurse and con-tact the anesthesiologist, who clears the patient if appropri-ate. Target time frames are included in the SOP.

SIMULATION MODELING

Complex simulation models combining discrete event sim-ulation (DES) and agent-based simulation (ABS) were builtto help prioritize initiatives intended to address overarch-

6 Erin M. Kane, MD, et al. Use of Systems Engineering to Design a Hospital Command Center

latipsoH

Capacity Optimization Governance Structure

Capacity Management Committee(Directors of Nursing, Administrators, Vice Chairs)

Capacity Optimization Executive Committee(President, Chief Operating Officer, Chief Nursing Officer, Vice President of Medical Affairs, Chief Administrative Officer Capacity Mgmt, Director of

Patient/Family and Visitor Services, Medical Director)

Office of Capacity Management(Chief Administrative Officer Capacity Mgmt, Director of Patient/Family

and Visitor Services, Medical Director, Assistant Medical Director, Assistant Director of Bed Management, Senior Project Administrator

Clinical Directors

/ lanoitcnuF

latnemtrape

D

Command Center

Nursing

Care Coordination

Perioperative

Emergency Department

Providers

Essential Services

Directors of Nursing Standard of Practice

Functional Units

Operating Room Executive Committee

Clinical Operations Committee Clinical Executive Committee

Clinical Directors Medical Board

Advisory Steering Huddles Command Center Operations

Perioperative Leadership

Leadership

Figure 3: The governance structure for capacity optimization is shown in this organizational chart. The core leadership

group for the Capacity Command Center meets weekly and drives strategic and operational initiatives.

ing capacity goals. DES is a computerized method of imi-tating the operation of a process over time. In health care,DES often focuses on the allocation of physical and hu-man resources and its effect on patient flow. These DESmodels are comprised of entities (for example, patients,laboratory tests) that flow through the modeled system,resources that process entities, locations, and arrival andprocessing interval distributions. 48 Alternatively, ABS is acomputational method of imitating the behavior of enti-ties (for example, patients, physicians) and interactions be-tween them. ABS in health care commonly focuses on howpeople make decisions within a process. ABS is particu-larly helpful when health care delivery process changes area function of the entities’ behavior, rather than an inputof the model (that is, agents decide what process to follownext). 49

In brainstorming potential solutions to our capacitychallenges, teams considered a wide variety of options (fullystaffed ICUs, additional monitored beds, timely hospitaldischarge, implementation of an extended stay unit, thecreation of a pediatric observation unit, decreasing PACUlength of stay, and many others). Simulation modeling wasused to predict the impact of each. With this quantificationof expected impact, the slate of initiatives could be moreintelligently prioritized. For instance, one of the multiplescenarios tested in the simulations was the shifting of 13

underutilized Medicine beds from an older section of the

hospital to a newer area. The work group hypothesized thatmoving these beds would have a significant impact on EDboarding due to shorter transport times and easier patientflow, but simulation modeling showed minimal effects onboarding times, so other initiatives were prioritized instead,including adjusting Medicine staffing to better match ad-mission requests and reducing “bed downtime” between in-patient admissions.

Technical aspects of these simulation models have beendescribed elsewhere. 50,51 Use of simulation modeling is keyto our ongoing operations.

Governance

The governance structure for the CCC has been essentialto our success. Planning for the CCC began in the summerof 2014 as an extension of a previously existing, multi-disciplinary Patient Flow Workgroup. With consultativesupport from GE Healthcare, the JHH placed the projectunder the administrative leadership of the Chief Adminis-trative Officer of Capacity Management and the Directorof Patient and Visitor Services. A steering committee forthe project was formed and includes the Johns HopkinsHealth System CEO, Chief Information Officer (CIO),and chief operating officer (COO), the JHH President andVice President for Medical Affairs, the JHH Director ofNursing, selected School of Medicine department directors,and the CCC project leaders.

Volume 000, No. , January 2019 7

Figure 4: This timeline illustrates key milestones in the development of the Capacity Command Center at the Johns Hop- kins Hospital.

As one of its first tasks, the steering committee definedthe role of the CCC medical director. The notion of a med-ical director has particular salience at an academic medicalcenter where the physician staff are employed by a School ofMedicine independent of the hospital. For this reason, theCCC must operate with lines of accountability that extendacross both the university and the health system. A medi-cal director with a faculty appointment serves as a liaisonto department directors and a leader for physician change.The medical director shapes strategic and operational de-cisions, bringing frontline perspective and highlighting po-tential downstream effects on patient care. This role alsoengages in day-to-day decisions in the CCC, such as adju-dicating disputed patient bed assignments and participatingin discussions about complex transfers.

A core leadership group for the CCC meets weekly anddrives strategic and operational initiatives. It includes theadministrative and medical leadership of the CCC as wella dedicated data analytics team and a group of clinical de-partment representatives ( Figure 3 ). The capacity manage-ment leadership team now reports directly to the executiveleadership of the organization, meeting on a regular basis to

discuss issues of capacity, patient flow, and patient demandpatterns. Capacity management is no longer a project but isnow a function within the hospital organizational structure.

PLANNED EVALUATION AND NEXT STEPS

Over the past several years, we’ve built infrastructure andcapabilities to actively manage capacity. Key milestones andpivots are highlighted in Figure 4 . The primary focus ofthe CCC continues to be expediting patient care throughoperational excellence.

Early data are encouraging. At the highest level, we’veimproved subcycle times and allowed our flagship hospitalto increase from 85% to 92% occupancy while decreasingpatient delays. For instance, ED boarding with Departmentof Medicine inpatient beds at 92% occupancy was previ-ously 9.7 hours and is now 6.3 hours ( Figure 5 ). Dischargesbefore noon increased modestly in Medicine and Neuro-sciences (Appendix 1, available in online article). Likewise,since the launch of a major initiative in November 2017,the number of OR holds has steadily decreased in the con-text of a constant case volume (Appendix 2, available in on-

8 Erin M. Kane, MD, et al. Use of Systems Engineering to Design a Hospital Command Center

Figure 5: This chart shows the operational occupancy (occupied beds/available beds) vs. median boarding hours (time

from request for admission bed to ED departure) for patients admitted to the Department of Medicine. Average operational occupancy for the Department of Medicine at the Johns Hopkins Hospital in fiscal year 2018 was 95%. Inpatient length of stay for the Department of Medicine during these three periods was relatively constant to increased at 5.3 days, 5.5 days, and 5.7 days, respectively. ED, emergency department.

line article). Though our demand still exceeds our capacity,it is estimated that the capacity optimization efforts haveadded the equivalent capacity of opening 13 to 15 addi-tional beds. It is difficult to know which elements of thiseffort have had the strongest impact, but transparency, visi-bility, and accountability for subcycle times and operationalmetrics have been paramount.

The next frontiers for the Judy Reitz Capacity Com-mand Center at the JHH are (1) channeling the centralizedexpertise and data into the creation of a daily schedule forinpatients, (2) developing additional tools to optimize re-source utilization and productivity, (3) pioneering the inte-gration of predictive clinical analytics into the CCC, and (4)strategically harmonizing specialty capacity and program-ming across the health system. Efforts are also ongoing todistribute the powerful tiles on the Wall of Analytics toother areas of the hospital.

CONCLUSION

We describe here the design and elements of the Judy ReitzCapacity Command Center. Key elements borrowed fromcommand centers in other industries include strategic colo-cation of teams, automated visual displays of real-time data

providing a global view, predictive analytics, standard workand rules-based protocols, and a robust governance struc-ture with guiding tenets. We built our command centerwith three primary goals—reducing ED boarding, elimi-nating OR holds, and facilitating transfers in from out-side facilities—but the command center infrastructure hasthe potential to improve hospital operations in many otherareas.

Acknowledgments. The authors wish to acknowledge the support of Car- rie Price, Welch Medical Library Clinical Informationist, in the preparation of this manuscript. Conflicts of Interest. All authors report no conflicts of interest.

SUPPLEMENTARY MATERIALS

Supplementary material associated with this article can befound, in the online version, at doi: 10.1016/j.jcjq.2018.11.006 .

Erin M. Kane, MD , is Assistant Professor, Department of Emergency Medicine, George Washington University School of Medicine and Health Sciences, Washington, DC. James J. Scheulen, PA, MBA , is Chief Admin- istrative Officer, Emergency Medicine and Capacity Management, Johns Hopkins Medicine, Baltimore. Adrian Püttgen, MD , is Associate Medi-

Volume 000, No. , January 2019 9

cal Director for Hospital-Based Neurosciences, Intermountain Healthcare, Salt Lake City. Diego Martinez, PhD , is Assistant Professor, Department of Emergency Medicine, Johns Hopkins University School of Medicine, and Data Scientist, Department of Operations Integration, Johns Hopkins Hospital. Scott Levin, PhD , is Associate Professor and Assistant Director for Research, Department of Emergency Medicine, Johns Hopkins Univer- sity School of Medicine. Bree A. Bush is Vice President, GE Healthcare. Linda Huffman, DNP, RN , is Assistant Director of Central Bed Manage- ment, Johns Hopkins Hospital. Mary Margaret Jacobs, MS, RN , is Di- rector of Patient and Visitor Services, Johns Hopkins Hospital. Hetal Ru- pani, MHA , is Director of Strategic Initiatives and Business Development, Office of Johns Hopkins Physicians. David T. Efron, MD , is Chief, Di- vision of Acute Care Surgery, and Director, Adult Trauma, Department of Surgery, Johns Hopkins University School of Medicine. Please address correspondence to Erin M. Kane, erin.kane@post.harvard.edu .

REFERENCES

1. Moskop JC , et al. Emergency department crowding,part 1—concept, causes, and moral consequences. AnnEmerg Med. 2009;53:605–611 .

2. Rabin E , et al. Solutions to emergency department “board-ing” and crowding are underused and may need to be leg-islated. Health Aff (Millwood). 2012;31:1757–1766 .

3. Kelen GD , Scheulen JJ . Commentary: emergency depart-ment crowding as an ethical issue. Acad Emerg Med.2007;14:751–754 .

4. American College of Emergency Physicians. PolicyStatement: Crowding. Feb 2013. Accessed Dec 14,2018. https://www.acep.org/globalassets/new-pdfs/policy-statements/crowding.pdf.

5. Institute of Medicine. Hospital-Based Emergency Care: Atthe Breaking Point. Washington, DC: National AcademiesPress, 2006 .

6. Hoot NR , Aronsky D . Systematic review of emergency de-partment crowding: causes, effects, and solutions. AnnEmerg Med. 2008;52:126–136 .

7. Liu SW , et al. A pilot study examining undesirable eventsamong emergency department-boarded patients awaitinginpatient beds. Ann Emerg Med. 2009;54:381–385 .

8. Singer AJ , et al. The association between length of emer-gency department boarding and mortality. Acad EmergMed. 2011;18:1324–1329 .

9. Pines JM , Griffey RT . What we have learned from adecade of ED crowding research. Acad Emerg Med.2015;22:985–987 .

10. Sun BC , et al. Effect of emergency department crowd-ing on outcomes of admitted patients. Ann Emerg Med.2013;61:605–611.e6 .

11. Derose SF , et al. Emergency department crowding predictsadmission length-of-stay but not mortality in a large healthsystem. Med Care. 2014;52:602–611 .

12. Pines JM , et al. The effect of emergency department crowd-ing on patient satisfaction for admitted patients. AcadEmerg Med. 2008;15:825–831 .

13. Levin S , et al. Tracking workload in the emergency depart-ment. Hum Factors. 2006;48:526–539 .

14. Fordyce J , et al. Errors in a busy emergency department.Ann Emerg Med. 2003;42:324–333 .

15. Trzeciak S , Rivers EP . Emergency department overcrowdingin the United States: an emerging threat to patient safetyand public health. Emerg Med J. 2003;20:402–405 .

16. Stang AS , et al. Crowding measures associated with thequality of emergency department care: a systematic review.Acad Emerg Med. 2015;22:643–656 .

17. McGowan JE , et al. Operating room efficiency and hospitalcapacity: factors affecting operating room use during max-imum hospital census. J Am Coll Surg. 2007;204:865–871 .

18. Dexter F , et al. Strategies to reduce delays in admissioninto a postanesthesia care unit from operating rooms. J Pe-rianesth Nurs. 2005;20:92–102 .

19. Chalfin DB , et al. Impact of delayed transfer of critically illpatients from the emergency department to the intensivecare unit. Crit Care Med. 2007;35:1477–1483 .

20. Hager DN , et al. Intermediate care to intensive care triage: aquality improvement project to reduce mortality. J Crit Care.2017;42:282–288 .

21. Harris S , et al. Delay to admission to critical care andmortality among deteriorating ward patients in UK hospi-tals: a multicentre, prospective, observational cohort study.Lancet. 2015 Feb 25;385:S40 .

22. McHugh M, et al. Improving Patient Flow and Reduc-ing Emergency Department Crowding: A Guide for Hos-pitals. Agency for Healthcare Research and Quality. Oct2011. AHRQ Publication No. 11(12)-0094. Accessed Dec 14,2018 https://www.ahrq.gov/sites/default/files/publications/files/ptflowguide.pdf.

23. McClelland MS , et al. The past, present, and future ofurgent matters: lessons learned from a decade of emer-gency department flow improvement. Acad Emerg Med.2011;18:1392–1399 .

24. Wiler JL , et al. Emergency department performance mea-sures updates: proceedings of the 2014 Emergency De-partment Benchmarking Alliance Consensus Summit. AcadEmerg Med. 2015;22:542–553 .

25. National Academy of Engineering; Institute of Medicine.Building a Better Delivery System: A New Engineer-ing/Health Care Partnership. Washington, DC: NationalAcademies Press, 2005 .

26. President’s Council of Advisors on Science andTechnology. Better Health Care and Lower Costs:Accelerating Improvement Through Systems En-gineering. May 2014. Accessed Dec 14, 2018.https://obamawhitehouse.archives.gov/sites/default/files/ microsites/ostp/systems _ engineering _ and _ healthcare.pdf.

27. Southard PA , et al. Impact of a transfer center on interhos-pital referrals and transfers to a tertiary care center. AcadEmerg Med.. 2005;12:653–657 .

28. Centralized process facilitates transfers. Hosp Case Manag.2012;20:75–76 .

29. Tortorella F , et al. Improving bed turnover time with a bedmanagement system. J Nurs Adm. 2013;43:37–43 .

30. Hemphill R , Nole B . Relieving an overcrowded ED and in-creasing capacity for regional transfers: one hospital’s bedmanagement strategies. J Emerg Nurs. 2005;31:243–246 .

31. Lovett PB , Illg ML , Sweeney BE . A successful model for acomprehensive patient flow management center at an aca-demic health system. Am J Med Qual. 2016;31:246–255 .

32. Davenport PB , et al. Integrating high-reliability principles totransform access and throughput by creating a centralizedoperations center. J Nurs Adm. 2018;48:93–99 .

33. Morris MK , Carter KF . A blended transfer and communica-tions center: designing a state-of-the-art mission control.Nurs Adm Q. 2015;39:357–361 .

34. Jaludi AA . Command Center Handbook: Proactive IT Mon-itoring: Protecting Business Value Through Operational Ex-cellence. Scotts Valley, CA: CreateSpace, 2014 .

35. International Olympic Committee. Technology Op-erations Centre Opens for Rio 2016. Nov 28, 2015.Accessed Dec 14, 2018. http://www.olympic.org/news/technology- operations- centre- opens- for- rio- 2016 .

10 Erin M. Kane, MD, et al. Use of Systems Engineering to Design a Hospital Command Center

36. Singer N. Mission Control, Built for Cities: I.B.M.Takes “Smarter Cities” Concept to Rio de Janeiro.New York Times. Mar 3, 2012. Accessed Dec 14,2018 https://www.nytimes.com/2012/03/04/business/ibm- takes- smarter- cities- concept- to- rio- de- janeiro.html .

37. Sridhar B , Grabbe SR , Mukherjee A . Modeling and opti-mization in traffic flow management. Proc IEEE Inst ElectrElectron Eng. 2008;96:2060–2080 .

38. Agrawal S. Why Hospitals Need Better Data Sci-ence. Harvard Business Review. Oct 19, 2017.Accessed Dec 14, 2018 https://hbr.org/2017/10/why- hospitals- need- better- data- science .

39. Herbsleb JD , et al. . Distance, dependencies, and delay ina global collaboration. CSCW Conf Comput Support CoopWork; 2000:319–328 .

40. Ginsburg S . Colocating health services: a way to improvecoordination of children’s health care? Issue Brief (Com-monw Fund). 2008;41:1–11 .

41. Fesko S, Timmons JC, Cohen A. Case Studies on the Im-plementation of the Workforce Investment Act: Focus onCo-location. Institute for Community Inclusion, 2003. Issue4Accessed Dec 14, 2018 https://www.communityinclusion.org/pdf/cs4.pdf.

42. Kraut RE , Streeter LA . Coordination in software develop-ment. Commun ACM. 1995;38(3):69–81 .

43. Herbsleb JD , Grinter RE . Splitting the organization and in-tegrating the code: Conway’s Law revisited. In: Proc Int ConfSoftw Eng.; 1999, 85–95 .

44. Porter ME , Heppelmann JE . How smart, connectedproducts are transforming companies. Harv Bus Rev.2015;93(10):96–112 114 .

45. Puget JF. Analytics Landscape. IBM Commu-nity, Sep 21, 2015. Accessed Dec 14, 2018 https://www.ibm.com/developerworks/community/blogs/jfp/ entry/Analytics _ Models?lang=en .

46. Lismont J , et al. Defining analytics maturity indicators: a sur-vey approach. Int J Inf Manage. 2017;37:114–124 .

47. Davenport TH , Harris JG . Competing on Analytics: The NewScience of Winning. Boston: Harvard Business School Press,2007 .

48. Hamrock E , et al. Discrete event simulation for healthcareorganizations: a tool for decision making. J Healthc Manag.2013;58:110–124 .

49. Siebers PO , et al. Discrete-event simulation is dead, longlive agent-based simulation! J Simul. 2010;4:204–210 .

50. Siddiqui S , Morse E , Levin S . Evaluating nurse staffing levelsin perianesthesia care units using discrete event simulation.IISE Trans Healthc Syst Eng. 2017;7:215–223 .

51. Levin S, Garifullin M. Simulating wait time in healthcare: ac-counting for transition process variability using survival anal-yses. In Yilmaz L., et al, editors: Proceedings of the 2015Winter Simulation Conference. Piscataway, NJ: IEEE Press,2015, 1252–1260.